From c1d626d9aa7a596aa7ec5ac50818d5f105f91353 Mon Sep 17 00:00:00 2001 From: Ray Liu Date: Wed, 23 Sep 2026 00:41:20 -0400 Subject: [PATCH] [6/6][load] publish shard-20 ramp run --- .../analysis.md | 3 + .../analysis.md | 3 + .../analysis.md | 3 + .../analysis.md | 121 ++++ .../result.json | 652 ++++++++++++++++++ .../run.json | 78 +++ .../analysis.md | 3 + .../analysis.md | 3 + 8 files changed, 866 insertions(+) create mode 100644 docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/analysis.md create mode 100644 docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/result.json create mode 100644 docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/run.json diff --git a/docs/performance/runs/2026-09-23-adhoc-lean-probe-capacity-33c95bf/analysis.md b/docs/performance/runs/2026-09-23-adhoc-lean-probe-capacity-33c95bf/analysis.md index 5343c4f..a23d6c9 100644 --- a/docs/performance/runs/2026-09-23-adhoc-lean-probe-capacity-33c95bf/analysis.md +++ b/docs/performance/runs/2026-09-23-adhoc-lean-probe-capacity-33c95bf/analysis.md @@ -25,6 +25,9 @@ This is the second of the runs published on 2026-09-23: 5. [Shard-10 ramp](../2026-09-23-adhoc-shard-10-ramp-33c95bf/analysis.md): ramp arrival with 1000 members x concurrency 10; 20,000 RPS is the highest rate near the Redis reference, limited by slot capacity. +6. [Shard-20 ramp](../2026-09-23-adhoc-shard-20-ramp-33c95bf/analysis.md): + concurrency 20 per member raised the ceiling to 34,000 RPS at 16 + partitions with queue time below the Redis reference. ## Purpose diff --git a/docs/performance/runs/2026-09-23-adhoc-queue-latency-diagnosis-33c95bf/analysis.md b/docs/performance/runs/2026-09-23-adhoc-queue-latency-diagnosis-33c95bf/analysis.md index b66bec0..41f869e 100644 --- a/docs/performance/runs/2026-09-23-adhoc-queue-latency-diagnosis-33c95bf/analysis.md +++ b/docs/performance/runs/2026-09-23-adhoc-queue-latency-diagnosis-33c95bf/analysis.md @@ -23,6 +23,9 @@ This is the third of the runs published on 2026-09-23: 5. [Shard-10 ramp](../2026-09-23-adhoc-shard-10-ramp-33c95bf/analysis.md): ramp arrival with 1000 members x concurrency 10; 20,000 RPS is the highest rate near the Redis reference, limited by slot capacity. +6. [Shard-20 ramp](../2026-09-23-adhoc-shard-20-ramp-33c95bf/analysis.md): + concurrency 20 per member raised the ceiling to 34,000 RPS at 16 + partitions with queue time below the Redis reference. ## Purpose diff --git a/docs/performance/runs/2026-09-23-adhoc-shard-10-ramp-33c95bf/analysis.md b/docs/performance/runs/2026-09-23-adhoc-shard-10-ramp-33c95bf/analysis.md index a7e0906..6e63a6c 100644 --- a/docs/performance/runs/2026-09-23-adhoc-shard-10-ramp-33c95bf/analysis.md +++ b/docs/performance/runs/2026-09-23-adhoc-shard-10-ramp-33c95bf/analysis.md @@ -26,6 +26,9 @@ This is the fifth of the runs published on 2026-09-23: 5. **This run.** Hold the run 4 topology fixed and ramp arrival at 8, 16, 32, and 64 partitions to find the highest rate that keeps queue time near the Redis reference. +6. [Shard-20 ramp](../2026-09-23-adhoc-shard-20-ramp-33c95bf/analysis.md): + concurrency 20 per member raised the ceiling to 34,000 RPS at 16 + partitions with queue time below the Redis reference. ## Purpose diff --git a/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/analysis.md b/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/analysis.md new file mode 100644 index 0000000..5fafddc --- /dev/null +++ b/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/analysis.md @@ -0,0 +1,121 @@ +# Run Analysis + +Run: [run.json](run.json) + +Results: [result.json](result.json) + +Kafka: [compose.yaml](../2026-09-23-adhoc-share-group-1000-members-33c95bf/compose.yaml) +(`group.share.max.size=1000`, `group.share.partition.max.record.locks=4000`) + +This document is generated by AI from the run metadata, machine results, and +retained diagnostics. + +## Sequence + +This is the sixth and last of the runs published on 2026-09-23: + +1. [Harness capacity](../2026-09-23-north-star-steady-v1-harness-capacity-33c95bf/analysis.md): + `kq-bench` saturated the host near 13,000 tasks/s. +2. [Lean-probe capacity](../2026-09-23-adhoc-lean-probe-capacity-33c95bf/analysis.md): + kq reached 150,000 tasks/s, but queue p50 stayed near 465 ms. +3. [Queue-latency diagnosis](../2026-09-23-adhoc-queue-latency-diagnosis-33c95bf/analysis.md): + queue time comes from the worker generation wait. +4. [Share group with 1000 members](../2026-09-23-adhoc-share-group-1000-members-33c95bf/analysis.md): + 1000 members x concurrency 10 sustained 20,000 RPS with low queue time at + 8 to 32 partitions and collapsed at 2 and 4. +5. [Shard-10 ramp](../2026-09-23-adhoc-shard-10-ramp-33c95bf/analysis.md): + 20,000 RPS was the highest rate near the Redis reference, limited by slot + capacity. +6. **This run.** Repeat the run 5 ramp with concurrency 20 per member to test + whether more slots per member raise that ceiling. + +## Purpose + +Find the highest arrival rate at which 1000 share-group members x concurrency +20 keep north-star queue time near the Redis reference (36 / 77 / 100 ms +p50 / p95 / p99), ramping arrival at 8, 16, 32, and 64 ready partitions with +the record-lock window fixed at 4000. + +The acceptance limits match run 5: every task completes exactly once, enqueue +reaches at least 99% of the requested rate, completions end within 3 s of +production, and queue p50 / p95 / p99 are at most 50 / 150 / 300 ms. + +## Observations + +| Partitions | Requested RPS | Median completions/s | Queue p50 / p95 / p99 / max | Outcome | +| ---: | ---: | ---: | ---: | --- | +| 8 | 20,000 | 19,991 | 4 / 13 / 22 / 577 ms | meets target | +| 8 | 24,000 | 23,994 | 5 / 19 / 141 / 826 ms | meets target | +| 8 | 26,000 | 26,018 | 5 / 49 / 273 / 767 ms | meets target | +| 8 | 28,000 | 28,120 | 7 / 192 / 417 / 1,147 ms | sustained; over limit | +| 16 | 24,000 | 24,000 | 8 / 28 / 45 / 768 ms | meets target | +| 16 | 28,000 | 28,008 | 10 / 35 / 53 / 678 ms | meets target | +| 16 | 32,000 | 32,000 | 16 / 51 / 73 / 483 ms | meets target | +| 16 | 34,000 | 33,984 | 22 / 66 / 95 / 802 ms | meets target | +| 16 | 36,000 | 36,022 | 49 / 155 / 217 / 558 ms | sustained; p95 over limit | +| 32 | 32,000 | 31,987 | 27 / 83 / 115 / 837 ms | meets target | +| 32 | 34,000 | 34,013 | 38 / 108 / 181 / 718 ms | meets target | +| 32 | 36,000 | 36,067 | 94 / 351 / 481 / 714 ms | sustained; over limit | +| 64 | 24,000 | 24,002 | 25 / 87 / 121 / 504 ms | meets target | +| 64 | 28,000 | 28,000 | 32 / 109 / 164 / 815 ms | meets target | +| 64 | 32,000 | 32,010 | 47 / 135 / 184 / 606 ms | meets target | +| 64 | 34,000 | 34,001 | 65 / 182 / 262 / 579 ms | sustained; over limit | +| 64 | 36,000 | 35,849 | 127 / 1,244 / 2,522 / 3,593 ms | sustained; tail 3 s | + +- Highest rate meeting the target: 26,000 RPS at 8 partitions, 34,000 at 16 + and 32, and 32,000 at 64. +- The best point was 34,000 RPS at 16 partitions with 22 / 66 / 95 ms queue + p50 / p95 / p99, below the Redis reference at every percentile. +- At 20,000 RPS and 8 partitions, concurrency 20 measured 4 / 13 / 22 ms + against 14 / 48 / 99 ms for concurrency 10 in run 5. +- No tested point overloaded outright. Completions tracked arrival up to the + highest tested rate of 36,000 RPS, where 64 partitions had the only + multi-second queue tail. +- At the same arrival rate, queue time rose with partition count above 16. +- All 30,480,000 tasks across all points completed with zero duplicates and + zero missing IDs. +- Process CPU at 34,000 to 36,000 RPS: Kafka 5.7 to 6.4 cores, producers 1.6 + to 1.9, workers 1.7 to 1.9, about 9.5 to 10.2 cores of the 12-core host + before other system processes. + +## Interpretation + +- Doubling per-member concurrency raised the latency-bounded ceiling from + 20,000 to 34,000 RPS and lowered queue time at equal load. Queue time is + governed mainly by spare slot capacity. Arrivals find an idle member sooner + when more slots are free, which outweighs the longer generation wait of 20 + records versus 10. This supersedes the earlier suggestion that smaller + generations are better. +- At 8 partitions, the limit near 26,000 to 28,000 RPS is consistent with the + in-flight window hypothesis from run 4: + `8 partitions x 4000 locks / about 1.1 s acquire-to-ack`. Not confirmed + from broker metrics. +- Above 16 partitions, latency again worsened with more partitions, as in + run 5. +- The throughput ceiling was not located; 36,000 RPS was the highest rate + tested. The host was near CPU saturation there, so this laptop cannot + separate a kq limit from a host limit above about 36,000 RPS. + +## Suggestions + +1. Continue on dedicated multi-core or cloud hardware, where Kafka, producers, + and workers do not share one host. +2. Test concurrency 40 and higher on that hardware to find where the longer + generation wait begins to outweigh added capacity. +3. Size partitions using the in-flight window estimate with headroom, and + avoid large surpluses of partitions. For this topology, 16 was the best + tested value. +4. Repeat the best points several times to bound run-to-run variance in p95 + and p99. + +## Caveats + +- Single Dockerized Kafka broker on localhost, RF=1; one run per point with a + 60 s arrival window. +- Same probe as runs 2 to 5: ready success path only, sleep-only handlers, + 200-byte payloads, queue time measured from just before `Enqueue`. +- The ramps for 16, 32, and 64 partitions started at 24,000 to 32,000 RPS, not + 20,000. +- The acceptance limits are an engineering judgment of "close to" the Redis + reference, not a production SLO. The Redis figures were not measured with + this workload or harness. diff --git a/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/result.json b/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/result.json new file mode 100644 index 0000000..4a74af1 --- /dev/null +++ b/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/result.json @@ -0,0 +1,652 @@ +{ + "correctness": { + "produced": 30480000, + "completed": 30480000, + "duplicates": 0, + "missing": 0 + }, + "summary_by_partitions": [ + { + "ready_partitions": 8, + "highest_rps_meeting_target": 26000, + "tested_rps": [ + 20000, + 24000, + 26000, + 28000 + ], + "maximum_median_completion_rps": 28120 + }, + { + "ready_partitions": 16, + "highest_rps_meeting_target": 34000, + "tested_rps": [ + 24000, + 28000, + 32000, + 34000, + 36000 + ], + "maximum_median_completion_rps": 36022 + }, + { + "ready_partitions": 32, + "highest_rps_meeting_target": 34000, + "tested_rps": [ + 32000, + 34000, + 36000 + ], + "maximum_median_completion_rps": 36067 + }, + { + "ready_partitions": 64, + "highest_rps_meeting_target": 32000, + "tested_rps": [ + 24000, + 28000, + 32000, + 34000, + 36000 + ], + "maximum_median_completion_rps": 35849 + } + ], + "points": [ + { + "label": "ramp20-p8-20k", + "requested_rps": 20000, + "duration_seconds": 60, + "ready_partitions": 8, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 20000, + "enqueue_errors": 0, + "completion_rps_median_1s": 19991, + "completion_rps_peak_1s": 20468, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 4, + "queue_p95_ms": 13, + "queue_p99_ms": 22, + "queue_max_ms": 577, + "correctness": { + "produced": 1200000, + "completed": 1200000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.06, + "producer": 1.42, + "workers": 1.66 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p8-24k", + "requested_rps": 24000, + "duration_seconds": 60, + "ready_partitions": 8, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 24000, + "enqueue_errors": 0, + "completion_rps_median_1s": 23994, + "completion_rps_peak_1s": 25631, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 5, + "queue_p95_ms": 19, + "queue_p99_ms": 141, + "queue_max_ms": 826, + "correctness": { + "produced": 1440000, + "completed": 1440000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.0, + "producer": 1.49, + "workers": 1.73 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p8-26k", + "requested_rps": 26000, + "duration_seconds": 60, + "ready_partitions": 8, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 26000, + "enqueue_errors": 0, + "completion_rps_median_1s": 26018, + "completion_rps_peak_1s": 26966, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 5, + "queue_p95_ms": 49, + "queue_p99_ms": 273, + "queue_max_ms": 767, + "correctness": { + "produced": 1560000, + "completed": 1560000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 5.97, + "producer": 1.54, + "workers": 1.77 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p8-28k", + "requested_rps": 28000, + "duration_seconds": 60, + "ready_partitions": 8, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 27999, + "enqueue_errors": 0, + "completion_rps_median_1s": 28120, + "completion_rps_peak_1s": 29863, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 7, + "queue_p95_ms": 192, + "queue_p99_ms": 417, + "queue_max_ms": 1147, + "correctness": { + "produced": 1680000, + "completed": 1680000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 5.96, + "producer": 1.62, + "workers": 1.76 + }, + "sustained": true, + "meets_latency_target": false + }, + { + "label": "ramp20-p16-24k", + "requested_rps": 24000, + "duration_seconds": 60, + "ready_partitions": 16, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 24000, + "enqueue_errors": 0, + "completion_rps_median_1s": 24000, + "completion_rps_peak_1s": 25980, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 8, + "queue_p95_ms": 28, + "queue_p99_ms": 45, + "queue_max_ms": 768, + "correctness": { + "produced": 1440000, + "completed": 1440000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.02, + "producer": 1.52, + "workers": 1.77 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p16-28k", + "requested_rps": 28000, + "duration_seconds": 60, + "ready_partitions": 16, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 28000, + "enqueue_errors": 0, + "completion_rps_median_1s": 28008, + "completion_rps_peak_1s": 28773, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 10, + "queue_p95_ms": 35, + "queue_p99_ms": 53, + "queue_max_ms": 678, + "correctness": { + "produced": 1680000, + "completed": 1680000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 5.89, + "producer": 1.55, + "workers": 1.79 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p16-32k", + "requested_rps": 32000, + "duration_seconds": 60, + "ready_partitions": 16, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 32000, + "enqueue_errors": 0, + "completion_rps_median_1s": 32000, + "completion_rps_peak_1s": 33519, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 16, + "queue_p95_ms": 51, + "queue_p99_ms": 73, + "queue_max_ms": 483, + "correctness": { + "produced": 1920000, + "completed": 1920000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 5.86, + "producer": 1.58, + "workers": 1.67 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p16-34k", + "requested_rps": 34000, + "duration_seconds": 60, + "ready_partitions": 16, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 33999, + "enqueue_errors": 0, + "completion_rps_median_1s": 33984, + "completion_rps_peak_1s": 36239, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 22, + "queue_p95_ms": 66, + "queue_p99_ms": 95, + "queue_max_ms": 802, + "correctness": { + "produced": 2040000, + "completed": 2040000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 5.69, + "producer": 1.58, + "workers": 1.66 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p16-36k", + "requested_rps": 36000, + "duration_seconds": 60, + "ready_partitions": 16, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 36000, + "enqueue_errors": 0, + "completion_rps_median_1s": 36022, + "completion_rps_peak_1s": 36759, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 49, + "queue_p95_ms": 155, + "queue_p99_ms": 217, + "queue_max_ms": 558, + "correctness": { + "produced": 2160000, + "completed": 2160000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.02, + "producer": 1.73, + "workers": 1.8 + }, + "sustained": true, + "meets_latency_target": false + }, + { + "label": "ramp20-p32-32k", + "requested_rps": 32000, + "duration_seconds": 60, + "ready_partitions": 32, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 31999, + "enqueue_errors": 0, + "completion_rps_median_1s": 31987, + "completion_rps_peak_1s": 32977, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 27, + "queue_p95_ms": 83, + "queue_p99_ms": 115, + "queue_max_ms": 837, + "correctness": { + "produced": 1920000, + "completed": 1920000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.06, + "producer": 1.7, + "workers": 1.7 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p32-34k", + "requested_rps": 34000, + "duration_seconds": 60, + "ready_partitions": 32, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 33999, + "enqueue_errors": 0, + "completion_rps_median_1s": 34013, + "completion_rps_peak_1s": 34442, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 38, + "queue_p95_ms": 108, + "queue_p99_ms": 181, + "queue_max_ms": 718, + "correctness": { + "produced": 2040000, + "completed": 2040000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.31, + "producer": 1.9, + "workers": 1.82 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p32-36k", + "requested_rps": 36000, + "duration_seconds": 60, + "ready_partitions": 32, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 35999, + "enqueue_errors": 0, + "completion_rps_median_1s": 36067, + "completion_rps_peak_1s": 36856, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 94, + "queue_p95_ms": 351, + "queue_p99_ms": 481, + "queue_max_ms": 714, + "correctness": { + "produced": 2160000, + "completed": 2160000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.05, + "producer": 1.74, + "workers": 1.75 + }, + "sustained": true, + "meets_latency_target": false + }, + { + "label": "ramp20-p64-24k", + "requested_rps": 24000, + "duration_seconds": 60, + "ready_partitions": 64, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 24000, + "enqueue_errors": 0, + "completion_rps_median_1s": 24002, + "completion_rps_peak_1s": 24177, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 25, + "queue_p95_ms": 87, + "queue_p99_ms": 121, + "queue_max_ms": 504, + "correctness": { + "produced": 1440000, + "completed": 1440000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.17, + "producer": 1.59, + "workers": 1.8 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p64-28k", + "requested_rps": 28000, + "duration_seconds": 60, + "ready_partitions": 64, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 3, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 28000, + "enqueue_errors": 0, + "completion_rps_median_1s": 28000, + "completion_rps_peak_1s": 33374, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 32, + "queue_p95_ms": 109, + "queue_p99_ms": 164, + "queue_max_ms": 815, + "correctness": { + "produced": 1680000, + "completed": 1680000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.1, + "producer": 1.61, + "workers": 1.72 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p64-32k", + "requested_rps": 32000, + "duration_seconds": 60, + "ready_partitions": 64, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 31999, + "enqueue_errors": 0, + "completion_rps_median_1s": 32010, + "completion_rps_peak_1s": 32611, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 47, + "queue_p95_ms": 135, + "queue_p99_ms": 184, + "queue_max_ms": 606, + "correctness": { + "produced": 1920000, + "completed": 1920000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.31, + "producer": 1.85, + "workers": 1.83 + }, + "sustained": true, + "meets_latency_target": true + }, + { + "label": "ramp20-p64-34k", + "requested_rps": 34000, + "duration_seconds": 60, + "ready_partitions": 64, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 33999, + "enqueue_errors": 0, + "completion_rps_median_1s": 34001, + "completion_rps_peak_1s": 34312, + "seconds_with_completions_after_production": 1, + "queue_p50_ms": 65, + "queue_p95_ms": 182, + "queue_p99_ms": 262, + "queue_max_ms": 579, + "correctness": { + "produced": 2040000, + "completed": 2040000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.31, + "producer": 1.88, + "workers": 1.85 + }, + "sustained": true, + "meets_latency_target": false + }, + { + "label": "ramp20-p64-36k", + "requested_rps": 36000, + "duration_seconds": 60, + "ready_partitions": 64, + "worker_processes": 10, + "workers_per_process": 100, + "worker_concurrency": 20, + "share_group_members": 1000, + "total_slots": 20000, + "producer_clients": 4, + "enqueue_goroutines_per_client": 512, + "achieved_enqueue_rps": 35999, + "enqueue_errors": 0, + "completion_rps_median_1s": 35849, + "completion_rps_peak_1s": 36339, + "seconds_with_completions_after_production": 3, + "queue_p50_ms": 127, + "queue_p95_ms": 1244, + "queue_p99_ms": 2522, + "queue_max_ms": 3593, + "correctness": { + "produced": 2160000, + "completed": 2160000, + "duplicates": 0, + "missing": 0 + }, + "cpu_cores_10s_window": { + "kafka_docker_vm": 6.42, + "producer": 1.92, + "workers": 1.84 + }, + "sustained": true, + "meets_latency_target": false + } + ] +} diff --git a/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/run.json b/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/run.json new file mode 100644 index 0000000..d905cb5 --- /dev/null +++ b/docs/performance/runs/2026-09-23-adhoc-shard-20-ramp-33c95bf/run.json @@ -0,0 +1,78 @@ +{ + "run_id": "2026-09-23-adhoc-shard-20-ramp-33c95bf", + "created_at": "2026-09-23T02:35:00Z", + "revision": "33c95bf74b5c", + "purpose": "Find the highest arrival rate at which 1000 share-group members x concurrency 20 keep north-star queue time near the Redis reference (36 / 77 / 100 ms), ramping arrival at 8, 16, 32, and 64 ready partitions with the record-lock window fixed at 4000.", + "scenario": null, + "profile": null, + "runner": "../2026-09-23-adhoc-lean-probe-capacity-33c95bf/probe/main.go via probe/run.sh with COMPOSE=../2026-09-23-adhoc-share-group-1000-members-33c95bf/compose.yaml SETTLE=40", + "command": "COMPOSE=../2026-09-23-adhoc-share-group-1000-members-33c95bf/compose.yaml SETTLE=40 ../2026-09-23-adhoc-lean-probe-capacity-33c95bf/probe/run.sh